Noisy fans

[akis]

I have been using a couple of 40mm fans inside a portable device to try and keep it cool. This device is on all the time, even in the night when everything else is very quiet, so the two fans must be as quiet as possible.

I bought a bunch of 40mm "Xilence" fans from here : http://www.amazon.co.uk/s/ref=nb_sb_noss?url=search-alias%3Daps&field-keywords=xilence+40mm&rh=i%3Aaps%2Ck%3Axilence+40mm

These fans spin slowly, drawing only 40mA at 12V rather than the typical 160mA of other 40mm fans which are much more powerful and noisy.

The problem is that after a few days of use, these fans start to make grinding noises. To demonstrate, if you power up the fan while holding it with two fingers, you can feel a vibration and then if you turn the fan in different positions you can actually hear the vibration too. When the fan is mounted on a plastic enclosure the plastic surfaces act as soundboards for those vibrations and it is even more loud. You can minimise the noise by turning the whole device in one direction or another, but it does not cure it completely and it is not practical having the device at some weird angle and position in order to keep it more quiet.

I have encountered this problem of the fan being noisy as if something is grinding inside and being even louder when turned to face a particular direction in other fans, so the problem is not limited to these fans.

Can someone recommend a make of fan I can get that will work 24/7 and not start to make noises in any orientation?

[rolycat]

Try using fans with better quality bearings. A quick search turned up these which use quiet and durable hydrodynamic bearings.

[wraper]

This is because rotor can play and vibrate significantly along the shaft axis in sleeve bearing fans. If you want that fan works well in any direction, then buy ones with ball bearings, not with sleeve bearings like xilence. Sunon maglev should perform well too.

[con-f-use]

I hate it when my fans are noisy. At least, they could have the decency to shut the hell up when stalking me. Damn groupies!



Joking aside, if you only want to cool one certain chip and energy efficiency is not a big deal, you could look into peltier elements.

[nctnico]

Peltier elements just make matters worse. As a rule of thumb they consume as much energy as they transport so doubling the cooling requirement. It makes more sense to remodel the case for a larger heatsink and natural convection. Maybe in combination with a heat pipe.

BTW: cheap fans are usually overpriced crap.

[Richard Head]

Get a Papst/EBM fan. They are twice the price but they are not Chinese shit that fails prematurely. They also specify the fan lifetime, not like the other makes I've seen.
You should get about 40 000h life from a 40 x 40 Papst fan.
Also, bigger fans last longer. I'm not sure why but that's just the way it is.
And NO, I don't work for them.

[a210210200]

For some actual quality overpriced stuff with a strange color scheme.

http://www.noctua.at/main.php?show=productview&setlng=en&products_id=47

I actually like noctua stuff it works fine but it is very expensive and has a lot of marketing crap but it works and their accessories are actually useful and their CPU heatsinks are so easy to install. I use them and scythe gentle typhoon fans. At 40mm though most fans become very similar and a good quality sleeve bearing should work fine for quite some time but a fluid dynamic/double ball bearing/maglev/.../... also work well.

Not sure if it really applies to computer fans but a fluid dynamic bearing does not like to start and stop or be shocked/bumped/hit depending on their design.

I've never used 40mm fans as they don't push much air and are generally noisy or if quiet don't push much air. A low profile blower or something might work as well.

The top quality stuff is from Sanyo Denki san-ace fans.

[miguelvp]

I hate it when my fans are noisy. At least, they could have the decency to shut the hell up when stalking me. Damn groupies!

So my wife likes to sleep with a fan on, and of course I don't, I did tell her I'm her biggest fan, but that didn't solve it,  I still have to deal with the fan noise while sleeping.

So a fan on my equipment is not going to bother me as much as those box fans.

[con-f-use]

Peltier elements just make matters worse. As a rule of thumb they consume as much energy as they transport so doubling the cooling requirement.

Not arguing with you, but in some applications they make sense. E.g where you only want to cool a small spot and transport the heat to the case or another passive cooler.

Also, bigger fans last longer. I'm not sure why but that's just the way it is.

Smaller fans spin faster and have less material in bearing that needs to meet tighter tolerances. So time till abrasion will make it fail is shorter.

[coppice]

Get a Papst/EBM fan. They are twice the price but they are not Chinese shit that fails prematurely. They also specify the fan lifetime, not like the other makes I've seen.
You should get about 40 000h life from a 40 x 40 Papst fan.
Also, bigger fans last longer. I'm not sure why but that's just the way it is.
And NO, I don't work for them.

He asked for quiet fans. Have you ever seen (heard)) a quiet Pabst fan?

[Richard Head]

Coppice
I must say I've never really noticed that the Papst fans are noisier. Perhaps they are, I'm not sure. What I do know is that as the required fan flow rate goes up so too does the noise.
I have designed equipment where the Papst fans (92 x 92) have been running for 10y continuously 24h/day without failure. That's not just a few but thousands in the field.
I think for high value applications they are definately worth it but consumer goods perhaps not.
Dick

[nctnico]

Peltier elements just make matters worse. As a rule of thumb they consume as much energy as they transport so doubling the cooling requirement.

Not arguing with you, but in some applications they make sense. E.g where you only want to cool a small spot and transport the heat to the case or another passive cooler.

No, that is what a heat pipe is for. With a peltier you cool things that need to get colder than the room temperature 

[coppice]

Coppice
I must say I've never really noticed that the Papst fans are noisier. Perhaps they are, I'm not sure. What I do know is that as the required fan flow rate goes up so too does the noise.
I have designed equipment where the Papst fans (92 x 92) have been running for 10y continuously 24h/day without failure. That's not just a few but thousands in the field.
I think for high value applications they are definitely worth it but consumer goods perhaps not.
Dick

The OP didn't ask for high lifetime. He asked for quiet. Pabst certainly make fans that last, but I've never seen anything quiet from them, despite their making a range with fancy feathered edge blades that are specifically intended to be quiet. A lot of the better lifetime fans are pretty noisy. I used to use a lot of Rotrons that we could hear above the engines on planes, but they certainly lasted.

People focusing on low noise fans are mostly targeting domestic applications which won't bear the cost of high lifetime bearings.

[SeanB]

Low noise is easy at the expense of much reduced air flow. Use a 12V fan on 5V, or a 24V fan on 12V and it will be a lot quieter, but the air flow will be down a lot. If the reduced flow is acceptable then simply using a lower DC voltage works well. Panasonic fans are very quiet.

[a210210200]

Be careful with using 12V fans at 5V they might not start and if they don't have stall protection you can burn the motor out. (I don't know if there are fans without it but you can tell when they do when you stop the fan by hand and it has torque pulses as it tries to restart and stalls again and cuts the power and waits a bit)

Good quality fans can be used at low voltages and many fans can keep running at a low voltage but might not start at the voltage from a stand still. (With build up of dust/age it may fail to startup later as well)

It is best/ideal to use an already slow speed fan or PWM fan and force the duty cycle to 0% so that you don't need to adjust the voltage and it can have maximum startup torque while still spinning slower.

That being said most fans will work fine down to 7V but 5V may cause problems with some. Sanyo denki fans even have performance curves at reduced voltages.

[akis]

Thanks for all the comments. My device uses two of these small fans, one to pull air into the device and another fan that blows straight into the heatsink which is inside the device and the very hot air is expelled at the back. The device is on 24/7 and the fans are also on all the time, there is no need to check temperatures, it is always hotter than it should be. The device is on all the time and that means when the room is quiet (eg night time) the fans become very annoying. I have discovered that the PCB onto which one of the fans is mounted acts as a soundboard, amplifying the rumble of the fan. I have added paper washers to the bolts but it still amplifies the sound. Ideally I should have used rubber mountings so that the bolt does not touch the PCB at all. Nevertheless it is the quality of the fans that matter, because in the beginning those fans were dead quiet. Then they started making noise (rumble). I have ordered two of the Noctua fans and will try them to see if they are more reliably quiet.

I have also ordered a papst fan for my fridge-freezer to replace its noisy fan, but the papst is noisier still so I have not yet bothered to replace it inside the freezer.

[RichardF]

So my wife likes to sleep with a fan on

It's not good for one's hearing and it interferes with deep sleep.

Devices that produce soothing sounds in order to lull infants to sleep can be loud enough to damage their hearing, researchers reported Monday.

The noise of a white noise machine used to get babies to sleep could damage delicate ears and even hinder auditory development.

The idea for the study on infant sleep machines, as this line of products is called, came from a check Papsin made on a young child who had undergone surgery at Sick Kids.

A sleep machine was operating loudly when he walked into the patient’s room. The parents informed the startled Papsin that they had been told to use the machine by a sleep doula.

He left the room and fetched one of his students and a sound level meter.

“In the room she records 85 decibels — the sound of a car wash. In the kid’s ear! To mask the noisy ward, which was probably 85 decibels itself.”

“I said ‘I’m going to go study this. This is crazy.’”

His team bought as many different types of sleep machines as they could find, both from baby supply stores and online sources. They found 14 different models which projected a variety of sounds such as a heartbeat, rain, crashing waves and white noise.

All but one of the machines, at all three distances, exceeded the 50 decibel level, Papsin and his team found. The one exception registered 49.2 decibels — and then only at the 200 cm. distance.

In fact, at 30 cm. — in the crib — three devices emitted sounds over 85 decibels, with one hitting just under 93 decibels. At the 100 cm. distance all the machines registered at 60 decibels and upward. The sound levels ranged from 49.2 to 76.2 decibels at the 200 cm. distance.

Papsin noted that these sound levels were in addition to whatever noise they were being used to mask.

It's not just infants whose hearing is damaged. People need quiet to rest their ear hairs and to get into deep enough sleep. The same goes for light. The retinas need to rest and light, even a small amount, interferes with deep sleep.

[Tallie]

I like Enermax's Marathon enlobal bearing fans. I've tried just about all the big names and they're the quietest and longest lasting. I don't think they come in 40mm, though.

[con-f-use]

It's not just infants whose hearing is damaged. People need quiet to rest their ear hairs [...]. The same goes for light. The retinas need to rest [...].

Quotation needed. I do NOT doubt that light and noise will interfere with your sleep. However, I'd need a scientific paper to believe that small amounts of sound or light will damage your sensory organs. Your blood, pillow movements and bodily functions will produce noise of similar intensity in your ears as a fan a few feet away. Your brain can adapt to small amounts of noise and light and there are many regions on earth where primates and humans lived that are noisy at night (e.g. the jungle).

[miguelvp]

It's not just infants whose hearing is damaged. People need quiet to rest their ear hairs [...]. The same goes for light. The retinas need to rest [...].

Quotation needed. I do NOT doubt that light and noise will interfere with your sleep. However, I'd need a scientific paper to believe that small amounts of sound or light will damage your sensory organs. Your blood, pillow movements and bodily functions will produce noise of similar intensity in your ears as a fan a few feet away. Your brain can adapt to small amounts of noise and light and there are many regions on earth where primates and humans lived that are noisy at night (e.g. the jungle).

Same here, need quotation, so I can convince my wife not to turn the damn fan on at night